A wide variety of processes ranging from enzyme catalysis, to oxygen binding in myoglobin, to energy and electron transfer in photosynthesis all rely on similar protein-cofactor behavior - the protein and cofactor combine to form a complex that is rigid on short timescales, yet flexible on long timescales. For example, in photosynthetic light harvesting complexes (LHCs), photon energy is retained in the pigments (not lost to protein motions) because the rigid system prevents protein reorganization during the pigment excited state lifetime (nanoseconds). However, the flexibility of the system on longer timescales varies the protein environment of each pigment molecule, yielding the broad absorption necessary for efficient light harvesting. Similarly, rigid short-time structure enables an enzyme to bind its substrate for optimum catalytic behavior and specificity of the reaction, while long-time flexibility allows release of the product and binding of a new substrate. Describing this combination of crystal-like short-time behavior and glass-like long-time behavior is crucial to understanding how proteins achieve efficient and specific biological function, and requires both structural and dynamical understanding of protein-cofactor interactions. Photosynthetic LHCs provide an ideal starting point for examination of this behavior. By using high-resolution crystal structures from several LHCs, ranging from plants to bacteria, general principles can be isolated from species peculiarities. Also, an enormous body of spectroscopic work exists for LHCs of all types, providing a strong basis for comparing simulation to experiment. MD simulations along with the MM-PBSA method and computational alanine scanning will reveal how individual residues and their motions affect various energies of the system, including protein- pigment interaction energies. Correlation of this data with the sequence and structure of the complex will reveal how specific motifs in the sequence and structure determine the unique dynamical properties of the complex. In addition, these results will show the contribution of specific residue interactions to the stability of the complex, addressing long-standing issues concerning differences in the behavior of some LHCs despite significant sequence homology and similar structures. A mixed MD/QM method will allow accurate calculations of the pigment transition energies permitting the MD simulation to be directly compared with spectroscopic data, and suggesting targets for site-directed mutagenesis studies. By elucidating the protein interactions critical for efficient light harvesting behavior, the proposed research seeks to identify general design principles used by nature to create protein-cofactor complexes of remarkably specific and efficient activity. Identifying these general principles is necessary to understand in detail, and to control, the activity of all proteins, including those important to human health.